Classification of the Stage of Dryness of Copra via Computer Vision Using Image Processing Techniques

The goal of this research is to establish a system that employs copra as a raw material to evaluate and classify its dryness. This will be achieved through a prototype designed with advanced image processing techniques. The system employs the YOLO v8 architecture and captures images of Copra samples with a Raspberry Pi Camera. Researchers have successfully created a working prototype to detect and classify the captured images into three grades of Copra dryness. The prototype accurately detects and classifies Copra samples based on the collected data. This advancement can potentially transform the copra industry by accelerating the selection process for the appropriate grade for raw material processing. By incorporating YOLO v8 as a fundamental component, this application adds valuable functionality to its library. The researchers recommend expanding the scope of the study by increasing the variety of raw material samples and incorporating additional products beyond Copra. The overall accuracy of the test conducted is eighty six percent with an error value of fourteen percent.